So, you heard that deliberate slip there, you know, she was going to forget about neuropathy, nephropathy and retinopathy. So yes, I am going to speak on neuropathy and retinopathy and I will speak to, a lot to neuropathy because that's my area of interest. But you will see that I also have, I think there are good things about retinopathy and the very good thing about retinopathy actually is that you can actually diagnose retinopathy very easily and either through digital retinal fundus photography or of course OCT or OCTA. So there are very good ways of detecting early diabetic retinopathy and doing something about it. Why is neuropathy the Cinderella complication even though there are lots of consequences of this? Well, that's because we, the American Diabetes Association included, still rely on this which is basically, sorry, yeah, the monofilament which I believe is a good test for identifying people at high risk of foot ulceration. However, it is not a good test for detecting early neuropathy and it certainly is not a test for painful neuropathy. And we do it, actually we do a bad job in terms of detecting diabetic peripheral neuropathy. So this is a recent review that we did and essentially what it shows is that across the world we are doing a terrible job in terms of diagnosing diabetic neuropathy. So this is important to note, percentage of undiagnosed diabetic painful neuropathy, whether it's Malaysia, Qatar, we actually showed this about 80%, US, 79% of people are still undiagnosed with this condition. And we have to ask ourselves the question why that is the case. And what are the consequences of not diagnosing this condition? One major consequence is, as you will know, this. So this is Mr. von Bergman in Berlin in 1901 and I can tell you not very much has changed between 1901 and 2024. Our vascular surgeons are still doing the same kind of operation and they are doing this probably more frequently now than they were before. Actually not probably, they are. There was a recent article in the New England Journal of Medicine that shows that amputations are on the increase. So what is the primary issue? And the primary issue for me is this. Neuropathy is, this is Machiavelli, as the doctors say the wasting disease to start with it is easy to cure but difficult to diagnose. After a time, unless it has been diagnosed and treated at the outset, it becomes easy to diagnose but difficult to cure. So if we are going to diagnose diabetic neuropathy when it's at an advanced stage, no matter what you do, which risk factor you treat, you will never ever be able to even stop progression or reverse that condition. And so we have to change the way we diagnose diabetic neuropathy. So currently with the tests that we have, whether it's monofilament, neurological exam, vibration perception, even nerve conduction, we are actually diagnosing this condition at this stage pretty much. But we know that the disease itself actually forms at least 20 years before this. So why are we not diagnosing the condition at an earlier stage? Part of the reason is we don't have sufficiently sensitive tests to be able to achieve this. So first step is going to have to be a mind, a change in the way we approach diabetic neuropathy. And we're going to have to remove these current tests. Are there other better tests to diagnose early diabetic neuropathy? Actually a questionnaire, a simple questionnaire would be sufficient to detect painful diabetic neuropathy. But I'm going to actually share with you my favorite topic, which is, because I'm talking about retinopathy as well, something called corneal confocal microscopy. So this is an eye test for neuropathy, not retinopathy, neuropathy. And the big difference is, this is an eye test for neuropathy. Why? Well, when you have a digital fundus camera, you look at the retina, here you're looking at the cornea. And the cornea, as it happens, is the most sensitive tissue in the body. Why is it the most sensitive tissue in the body? Well, I can just tell you, if you do that, it's okay. What about if you get a little bit of dirt, anything, particle in your eye, it's horrible. And actually, anatomically, if you look at the number of nerves per millimeter squared, there's cornea as the most densely innervated tissue. Now, does that mean that you have to take a bit of cornea out? Who's volunteering for a corneal biopsy? Anybody? No? Well, I've got an alternative for you, which is corneal confocal microscopy. And really, what I want to point out is, this is a study actually we did now over 20 years ago, where we took patients with diabetic neuropathy, mild, moderate, severe, based on nerve conductions, vibration perception, and neurological exam, we graded them into mild, moderate, and severe neuropathy. We then undertook corneal confocal microscopy. This is a healthy control. You can see multiple nerves. Here is a patient with severe diabetic neuropathy, you can see very few nerves, very few branches. And what you see is a progressive fallout of nerve fibers. But the most important is this, the early stages of diabetic neuropathy, when actually most of these tests are still normal, there is already a loss of nerve fibers. So that was 18 patients. But subsequently, actually, there are now multiple studies, 38 studies, over 4,000 patients that show exactly the same. This is a meta-analysis that shows you that corneal confocal microscopy differentiates healthy controls from people with diabetic neuropathy, and also even subclinical diabetic neuropathy, that is early, mild, and moderate neuropathy. And hopefully I'm gonna put this story together and merge some of the ideas that we are currently using in the world of retinopathy with neuropathy. And what is the one big thing that's happened in retinopathy recently? That is the application of AI to the assessment of retinal imaging. And actually, for neuropathy, we also now have used AI to look at these images. And what we show is that if you use, take these images, you put them into these deep learning models, then actually, with 100% sensitivity and 95% specificity, I can detect people with diabetic neuropathy. And even those with and without neuropathy, I can detect with 92% and 80% sensitivity and specificity. So I am out of a job. The computer doesn't need me. All you have to do is get the patient to have a scan, the AI will work out whether you do or you don't have diabetic neuropathy. Whether we like it or not, whether we accept it or not, this is gonna happen. And I believe that the future will be something like you've got a bunch of people, maybe we'll focus on diabetic patients, they will go to a center, they will have a corneal confocal microscopy, AI will come along, it will say, you are healthy, you don't have diabetic neuropathy. You have various forms of diabetic neuropathy. Then AI again, actually, and we've shown that now, further differentiates into people with pain and without pain. And really, there is no involvement from me. This is all done in an agnostic way. So once we've reached this stage, can we do anything about this? So this is an editorial in the Lancet of, well, 30 odd years ago. Basically, it was, I guess, echoing what most people think about neurologists, which is that you make the diagnosis, or they used to make the diagnosis, clever diagnosis, and then what do you do for it? Most diseases, this was 30 years ago. Things have changed. But so is there anything we can do for diabetic neuropathy, or do we just diagnose it? Well, I think there are things, actually, and that's where I think, again, our mindset has got to change. We have this nihilistic view that when you get diabetic neuropathy, when nerves are damaged, they're degenerated, there's nothing we can do, really. Some people might think glycemic control is important, but actually, most people think, really, I can give something for pain relief, maybe duloxetine, maybe gabapentin, but there's not much I can do for the underlying disease. What I would say to you is that we can do something about this, and that something we can do is based on, this is one study that we've done in the Middle East, actually, which is relevant to the region, but what it shows you is that there are certain risk factors for diabetic neuropathy. What are they? High glucose, hypertension, hyperlipidemia, being overweight, particularly, being less physically active, and smoking cigarettes, so here, automatically, you've got treatable risk factors for diabetic neuropathy. Now, what drugs do we have that control glucose and weight? Well, we have GLP-1 therapies, and in fact, there is a study, a sub-study, from the LEADER trial. This was actually liraglutide, which I think is a relatively poor GLP-1, but essentially, what it did is, in the LEADER trial, it was a cardiovascular outcome trial, but they looked at amputation and ulceration. They did not show a difference in terms of the number of ulcers that developed, so very comparable. However, the big difference arose when they looked at the people who had ulcers who went on to have amputation. This was liraglutide arm, this was the placebo arm. There was a significant reduction of about 35% in the number of amputations in the people on liraglutide compared to placebo, indicating that there's something going on with people on GLP-1 therapy that is perhaps preventing the big major outcome that we all don't want in our diabetic patients, and that could be neuropathy. And so, in fact, we, in Qatar, we did a study where we looked at whether or not GLP-1 therapy with pioglitazone or basal bolus insulin, which both improve glucose, can actually have an impact on neuropathy, and we did that using corneal confocal microscopy. So, first of all, what you see is patients with type 2 diabetes who are very poorly controlled, HbA1c of 10.5%, both arms, whether it was exenatide with pioglitazone or basal bolus, reduced their HbA1c to around 7%. We then undertook, over 12 months, corneal confocal microscopy at baseline and at 12 months in both arms, and what we see is the following, that because of the improvement in glycemic control, with GLP-1 or with basal bolus insulin, actually you had an improvement in corneal nerves. So three parameters, fiber density, branch density, and length, all improved significantly. We now actually have an ongoing study with semaglutide in overweight individuals, where you can see, with semaglutide, there is clearly an impact on weight, okay, about 10 kilograms weight loss. Improvement in HbA1c, improvement in triglycerides. We now are seeing, it's not significant yet, because this is about 60 patients, actually, but there is an improvement in terms of the DN4, which is the pain score. So there's an improvement in painful symptoms. But in addition to that, corneal nerves, which are now, we measured at three months, at six months, what you start to see is an increase in corneal nerve fiber density and length. So there's an improvement, but this is not yet significant. However, the branch density, which is the number of branches coming from the main trunk, which tells me about regeneration, actually improved significantly in patients at three months and at six months with semaglutide. We now have another study, actually, that is with darpagliflozin. This is with collaborators in India, where they took type two diabetic patients with diabetic neuropathy, and they randomized them to either darpagliflozin 10 milligrams, or standard of care, which was anything but darpagliflozin, or an SGLT2 inhibitor. You can see in both arms, there was a reduction in HbA1c by about 0.53%, and 0.57% in the standard care arm. Then, we looked, or they looked, and we measured this for them, their corneal nerves. And what you see is the following, that in the darpagliflozin arm, actually, in six months, there is already a significant increase in nerve fiber density, branch density, and length. The standard of care arm, for some reason, even though the HbA1c was equivalently reduced, didn't have the same effect. It wasn't significantly increased. So, maybe there's something special about SGLT2 inhibitors in neuropathy that is causing this regeneration of nerves. We don't know yet, but I can tell you, this study showed it. There is another study from India that is also showing the same thing. And there is a third study from Turkey, which is also showing the same thing. Next, obesity and physical, so I think I kind of dealt with obesity because GLP-1 and SGLT2s clearly work to lose weight as well as improve HbA1c. What about physical activity? Well, again, in Qatar, we have two kinds of people. There are those who do very little, and there are those who actually, and there are those who actually say they're active, but they do 30 minutes of walking more than three times a week. Now, that, to me, isn't active, but it's pretty good, okay? What we managed to do is take this cohort of two cohorts and actually measure their nerves before and after over two years. And what did we show? Well, basically, if you were active, your nerves got better. This is nerve fiber density. You can see in the blue is the active, physically active, increased, whereas in the physically inactive, you had your progressively lost nerve fibers, the red, whereas in the active arm, and this isn't that active, so what I would say to all of you, maybe some of you are more active, but 30 minutes of walking three times a week, and you will start to regenerate nerves, okay? Then we have hyperlipidemia. So hyperlipidemia, actually, the important thing here to note is that it's not high cholesterol. It's actually high triglycerides, which seem to be the major driver for neuropathy. And in fact, we have evidence, old evidence, this is actually published in 2009, showing that phenofibrate has this amazing effect on reducing amputation rate by 50%. So field study was actually a cardiovascular outcome trial, which was negative, but the investigators actually assessed within that cohort whether or not people had minor or major amputations. And what they showed is that there was a reduction of almost 50% in minor amputations, not major amputations. And that actually, at the time, I was thinking, how come not major, minor? Well, there's a reason for it. What do you think that might be? The reason is that which kind of patients have major amputations? Is it those with vascular disease or those with neuropathy? Vascular disease. Which kind of patients have minor amputations? Neuropathic, distal, small ulcers. And you can see, that to me suggested that it could be an effect on neuropathy. Now, that was 2009. In 2010, I went to Abbott and I said, look, you've got a drug that improves neuropathy. Did they fund it? No. However, we now have a study, actually from Singapore, with literally 30 type 2 diabetic patients who're given phenofibrate, 100 to 300 milligrams, for 30 days, just 30 days. What did they see? Well, first of all, there is a reduction in triglycerides from 1.8 to 1.4. But look at their corneal nerves. Within 30 days, you're starting to see repair, increase in nerve fiber density. So there is regeneration of nerves with this. What about diabetic retinopathy? Well, the most important thing that I think has happened for diabetic retinopathy is, what? Is it anti-VGF? Is it something else? No. The most important thing that's ever happened to diabetic retinopathy is screening. And that is a big bugbear I have. In this part of the world, where we have so much money, so much resources, yet we haven't got adequate, proper diabetic retinal screening programs. In England, which is bankrupt, the NHS, they recognized that if you look at diabetic retinopathy, vision only becomes involved when you get advanced changes. What you need to be doing is identifying patients when they still have normal vision, but they have mild stages of diabetic retinopathy. And Nicholas Wald actually worked out, this was the original paper, based on which the diabetic retinal screening program in England was started. 1989, they published a analysis which showed that if you have detectable cases, 88% you can detect. Treat, 87% you can treat. And you can prevent 73% of people going blind. And based on that argument, they actually instituted the diabetic retinal screening program, which was the English retinal screening program in 2003. What were the results of that? Well, very simple. If you look and compare, diabetic retinopathy, okay, was the leading cause of blindness in working age adults in England before the diabetic retinal screening program started. That was 1999. 2009, six years after the English retinal screening program, it is no longer the most common cause of blindness in working age adults. All they'd done is start to screen for diabetic patients. And then people said, and the conclusions were, for the first time in at least five decades, diabetic retinopathy and mycopathy is no longer the leading cause. But then one of the arguments always is, well, you need all these graders to sit. If you're doing all this retinal screening, you're gonna have to grade people and you're gonna have to employ, it's gonna be costly. But actually, that's all changed now. And I go back to artificial intelligence for DR screening. So what is artificial intelligence? Artificial intelligence essentially takes away the graders. This is a machine-based algorithm which will look at the images and will identify different grades of retinopathy, okay? And it will do so with high accuracy. So what you have is people come along. They don't even have to be dilated now. They will undergo retinal screening. The AI model will grade into those with and without sight-threatening retinopathy and you will then have appropriate treatment. So again, something that will literally take minutes once you have a digital fundus image, and it's only before there was lots of discussions about whether we have seven-field, 10-field, wide-field retinal screening. You just need one image. And from that one image, actually, now the FDA, they approved in 2018, the first of the AI algorithms. In 2022, IART was approved by the FDA for multiple cameras to identify vision-threatening diabetic retinopathy with a sensitivity of 96%, specificity of 91%. And just now in April 2024, this has been approved. The AIDS, this is a fully autonomous AI for DR screening. It's a handheld camera, which one image, one minute, 99% of people get a diagnosis with 92% sensitivity and 89% specificity. What about treatment? Well, for treatment, for diabetic retinopathy, clearly we have glycemic control, which has been shown to reduce retinopathy by 30%. And what do we see that is better than glycemic control? Phenofibrate actually had a better impact on reducing diabetic retinopathy, 40% as opposed to 37%. Blood pressure lowering actually didn't do that much, but there are, again, data from the field study, which was 200 milligrams of phenofibrate, which shows that macular edema, proliferative retinopathy and all retinopathy was significantly reduced over five years if you're on phenofibrate as opposed to placebo. And based on that, actually, in Australia, phenofibrate has a license for retinopathy. And there is now actually Abbott trying to get a worldwide license for retinopathy. There are also data, which are actually bigger population-based data, which have looked at people, 5,000 patients on phenofibrate as opposed to 144,000 not on phenofibrate. What do they show? Well, basically, vision-threatening diabetic retinopathy, significantly less, and proliferative diabetic retinopathy, significantly lower. Not macular edema, okay? But vision-threatening retinopathy and proliferative diabetic retinopathy was reduced in people on phenofibrate, but not for macular edema. Now, we actually have a randomized control trial just being published on the effects of phenofibrate called the LENS trial. And essentially, this is from Scotland, 1,100 participants with type 1 and type 2 diabetes on phenofibrate, 145 milligrams, okay? They were pretty much on all of the maximally controlled into the statins and ACE inhibitors. What did they show? Well, they showed that if you're on phenofibrate, you have a reduction in terms of referable diabetic retinopathy or maculopathy, significantly reduced, and progression of diabetic retinopathy or maculopathy. And last, we now also have data from the EMPIREG study showing that the SGLT2 inhibitors, empagliflozin may also have an impact, reduces diabetic retinopathy. And we have an analysis from actually Taiwan, which compared DPP4 inhibitors with SGLT2 inhibitors, pioglitazone with SGLT2 inhibitors, and sulfonylureas with SGLT2 inhibitors. What did they show? Well, the incidence of site-threatening retinopathy was significantly lower if you're on SGLT2 compared to sulfonylurea, compared to pioglitazone, and compared to DPP4 inhibitors. About a 40% lower incidence of site-threatening retinopathy if you're on SGLT2 as opposed to any of these other drugs. And last but not least, I'm finishing off now, GLP-1 and diabetic retinopathy. There is an elephant in the room, and the elephant in the room is the SUSTAIN-6 data, which suggested that semaglutide may increase the risk of diabetic retinopathy. Actually, if you look at the data, really, a lot's been made out of it, but it was only 50 in the semaglutide arm compared to 29 participants who had an event. There is no data apart from one recent prospective study which looked at GLP-1 against SGLT2. And what it essentially shows is that there is no signal for more retinopathy in the GLP-1 arm as opposed to the SGLT2. And we now have, of course, the FOCUS study, which is due to report in 2026, which is a 1,500-patient study with semaglutide, which is actually gonna address whether or not really there is an issue with retinopathy. So I'll finish there, and I'll tell you the eye is the window to the soul and beyond. Thank you.

diabetic neuropathy

retinopathy

AI screening

corneal confocal microscopy

GLP-1 therapies

SGLT2 inhibitors